PLSC30500, Fall 2021

# PLSC30500, Fall 2021
## 1.2 Data visualization

---

# Toward a grammar of graphics

---

class: bg-full
background-image: url("data:image/png;base64,#assets/rosling_youtube.png")
background-position: center
background-size: contain

???

Source: https://www.youtube.com/watch?v=jbkSRLYSojo

---

class: bg-full
background-image: url("data:image/png;base64,#assets/rosling_youtube_zoom.png")
background-position: center
background-size: contain

???

Source: https://www.youtube.com/watch?v=jbkSRLYSojo

---

# Mapping attributes to aesthetics

Q: What is the **unit of observation**?

A: A country in a year

Q: How are a country-year's **attributes** mapped to **aesthetic** components of the graphic?

<table>
 <thead>
  <tr>
   <th style="text-align:left;"> Attribute </th>
   <th style="text-align:left;"> Aesthetic </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> Income </td>
   <td style="text-align:left;"> Horizontal position (x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Life expectancy </td>
   <td style="text-align:left;"> Vertical position (y) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Population </td>
   <td style="text-align:left;"> Size of point </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Continent </td>
   <td style="text-align:left;"> Color of point </td>
  </tr>
</tbody>
</table>

---

class: bg-full
background-image: url("data:image/png;base64,#assets/Minard.png")
background-position: center
background-size: contain

### Minard's graphic on Napoléon in Russia

???

One of the "best statistical drawings ever created" (Tufte, *VDQI*)
Source: [Wikipedia](https://en.wikipedia.org/wiki/File:Minard.png)

---

# Mapping attributes to aesthetics

Q: What is the **unit of observation**?

A: An army (army division) on a day ("army-day")

Q: How are an army-day's **attributes** mapped to **aesthetic** components of the graphic?

<table>
 <thead>
  <tr>
   <th style="text-align:left;"> Attribute </th>
   <th style="text-align:left;"> Aesthetic </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> Longitude </td>
   <td style="text-align:left;"> Horizontal position (x) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Latitude </td>
   <td style="text-align:left;"> Vertical position (y) </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Number of surviving soldiers </td>
   <td style="text-align:left;"> Width of line </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Direction (advance, retreat) </td>
   <td style="text-align:left;"> Color of line </td>
  </tr>
</tbody>
</table>

(Also note secondary plot showing temperature during retreat.)

---

# Data: structure

Our data is typically **rectangular**, with rows and columns like a spreadsheet.

Usually,

- each row should be one observation (e.g. country-year, army-day)
- each column should contain one attribute (e.g. life expectancy, number of surviving troops)

For example:

<table class="table" style="font-size: 12px; margin-left: auto; margin-right: auto;">
 <thead>
  <tr>
   <th style="text-align:left;"> country </th>
   <th style="text-align:left;"> continent </th>
   <th style="text-align:right;"> lifeExp </th>
   <th style="text-align:right;"> pop </th>
   <th style="text-align:right;"> gdpPercap </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> Afghanistan </td>
   <td style="text-align:left;"> Asia </td>
   <td style="text-align:right;"> 43.828 </td>
   <td style="text-align:right;"> 31889923 </td>
   <td style="text-align:right;"> 974.5803 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Albania </td>
   <td style="text-align:left;"> Europe </td>
   <td style="text-align:right;"> 76.423 </td>
   <td style="text-align:right;"> 3600523 </td>
   <td style="text-align:right;"> 5937.0295 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Algeria </td>
   <td style="text-align:left;"> Africa </td>
   <td style="text-align:right;"> 72.301 </td>
   <td style="text-align:right;"> 33333216 </td>
   <td style="text-align:right;"> 6223.3675 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Angola </td>
   <td style="text-align:left;"> Africa </td>
   <td style="text-align:right;"> 42.731 </td>
   <td style="text-align:right;"> 12420476 </td>
   <td style="text-align:right;"> 4797.2313 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Argentina </td>
   <td style="text-align:left;"> Americas </td>
   <td style="text-align:right;"> 75.320 </td>
   <td style="text-align:right;"> 40301927 </td>
   <td style="text-align:right;"> 12779.3796 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Australia </td>
   <td style="text-align:left;"> Oceania </td>
   <td style="text-align:right;"> 81.235 </td>
   <td style="text-align:right;"> 20434176 </td>
   <td style="text-align:right;"> 34435.3674 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Austria </td>
   <td style="text-align:left;"> Europe </td>
   <td style="text-align:right;"> 79.829 </td>
   <td style="text-align:right;"> 8199783 </td>
   <td style="text-align:right;"> 36126.4927 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Bahrain </td>
   <td style="text-align:left;"> Asia </td>
   <td style="text-align:right;"> 75.635 </td>
   <td style="text-align:right;"> 708573 </td>
   <td style="text-align:right;"> 29796.0483 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Bangladesh </td>
   <td style="text-align:left;"> Asia </td>
   <td style="text-align:right;"> 64.062 </td>
   <td style="text-align:right;"> 150448339 </td>
   <td style="text-align:right;"> 1391.2538 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Belgium </td>
   <td style="text-align:left;"> Europe </td>
   <td style="text-align:right;"> 79.441 </td>
   <td style="text-align:right;"> 10392226 </td>
   <td style="text-align:right;"> 33692.6051 </td>
  </tr>
</tbody>
</table>

???

Data in this format is sometimes referred to as "tidy" [(Wickham 2014)](https://vita.had.co.nz/papers/tidy-data.pdf). I think this concept is useful as long as you recognize that the definition of "unit of observation" (and thus attribute/variable) depends on the purpose for which the data is being used.

---

# Making (beautiful and informative) graphics

---

# Making graphics in `R`

We will use the `ggplot2` library, which is part of the `tidyverse` library.

Basic components of plotting with `ggplot`:

- data 
- mapping of attributes (columns of data) to aesthetics
- geometric representations of data (`geom`s)

---

# Quick detour: getting data into `R`

---

# Getting data into `R` (and `RStudio`)

An interactive option: `Import Dataset` button in `Environment` pane of `RStudio`

But note it's showing you the code it's using!

(Live coding example.)

---

# Getting data into `R` (cont'd)

Most commonly used functions:

- `read_csv()` and `read_rds()` in `readr` (`tidyverse`)
- `readstata13::read.dta13()` for Stata files (`.dta`)
- `readxl::read_excel()` for Excel files (`.xls`, `.xlsx`)
- `load()` in base R for "R objects"

All require "path" to file as argument.

Sometimes data is a package, e.g. `babynames`, `gapminder`, `vdemdata`

See Chapter 11 of R4DS and "Data Import" cheatsheet.

---

## A simple example

Get data in:

```r
gapminder_2007 <- gapminder::gapminder %>% 
  filter(year == 2007 & continent != "Oceania")   # will explain next week
```

Make a plot:

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp)) + 
  geom_point()
```

---

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp)) + 
  geom_point()
```

To note:

- the **arguments** to `ggplot()` say what the data is (`data = gapminder_2007`) and how attributes are mapped to aesthetics (`mapping = aes(x = gdpPercap, y = lifeExp)`)
- `geom_point()` says "plot a point for each observation"
- **layers** of plot linked with plus sign (`+`)

---

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp,
*                    size = pop)) +
  geom_point()
```

---

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp,
                     size = pop,
*                    col = continent)) +
  geom_point()
```

---

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp,
                     size = pop,
                     col = continent)) +  
  geom_point() + 
* scale_x_log10()
```

---

# Minard data

<table class="table" style="font-size: 15px; margin-left: auto; margin-right: auto;">
 <thead>
  <tr>
   <th style="text-align:right;"> long </th>
   <th style="text-align:right;"> lat </th>
   <th style="text-align:right;"> survivors </th>
   <th style="text-align:left;"> direction </th>
   <th style="text-align:right;"> group </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:right;"> 24.0 </td>
   <td style="text-align:right;"> 54.9 </td>
   <td style="text-align:right;"> 340000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 24.5 </td>
   <td style="text-align:right;"> 55.0 </td>
   <td style="text-align:right;"> 340000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 25.5 </td>
   <td style="text-align:right;"> 54.5 </td>
   <td style="text-align:right;"> 340000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 26.0 </td>
   <td style="text-align:right;"> 54.7 </td>
   <td style="text-align:right;"> 320000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 27.0 </td>
   <td style="text-align:right;"> 54.8 </td>
   <td style="text-align:right;"> 300000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 28.0 </td>
   <td style="text-align:right;"> 54.9 </td>
   <td style="text-align:right;"> 280000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 28.5 </td>
   <td style="text-align:right;"> 55.0 </td>
   <td style="text-align:right;"> 240000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 29.0 </td>
   <td style="text-align:right;"> 55.1 </td>
   <td style="text-align:right;"> 210000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 30.0 </td>
   <td style="text-align:right;"> 55.2 </td>
   <td style="text-align:right;"> 180000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 30.3 </td>
   <td style="text-align:right;"> 55.3 </td>
   <td style="text-align:right;"> 175000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 32.0 </td>
   <td style="text-align:right;"> 54.8 </td>
   <td style="text-align:right;"> 145000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 33.2 </td>
   <td style="text-align:right;"> 54.9 </td>
   <td style="text-align:right;"> 140000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 34.4 </td>
   <td style="text-align:right;"> 55.5 </td>
   <td style="text-align:right;"> 127100 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 35.5 </td>
   <td style="text-align:right;"> 55.4 </td>
   <td style="text-align:right;"> 100000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 36.0 </td>
   <td style="text-align:right;"> 55.5 </td>
   <td style="text-align:right;"> 100000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 37.6 </td>
   <td style="text-align:right;"> 55.8 </td>
   <td style="text-align:right;"> 100000 </td>
   <td style="text-align:left;"> A </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 37.7 </td>
   <td style="text-align:right;"> 55.7 </td>
   <td style="text-align:right;"> 100000 </td>
   <td style="text-align:left;"> R </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
  <tr>
   <td style="text-align:right;"> 37.5 </td>
   <td style="text-align:right;"> 55.7 </td>
   <td style="text-align:right;"> 98000 </td>
   <td style="text-align:left;"> R </td>
   <td style="text-align:right;"> 1 </td>
  </tr>
</tbody>
</table>

---

```r
ggplot(data = minard,
       mapping = aes(x = long, y = lat, 
                     size = survivors, col = direction,
                     group = group)) + 
* geom_path()
```

![](data:image/png;base64,#slides_12_files/figure-html/unnamed-chunk-12-1.png)

???

"`geom_path()` connects the observations in the order in which they appear in the data. `geom_line()` connects them in order of the variable on the x axis. `geom_step()` creates a stairstep plot, highlighting exactly when changes occur."

Source: https://ggplot2.tidyverse.org/reference/geom_path.html

---

# Back to `gapminder`

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp,
                     size = pop, col = continent)) +  
  geom_point() +   
  scale_x_log10() 
```

---

# Adding a smoothing line

`geom_smooth()` adds a "smoother". Let's try adding it!

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp,
                     size = pop, col = continent)) +  
  geom_point() +   
  scale_x_log10() + 
* geom_smooth()
```

Hmm.

??? If you don't exclude Oceania, `ggplot` refuses to make a smoother using default settings because there are too few countries in Oceania.

---

# Inheritance of aesthetics

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp, size = pop)) +  
* geom_point(aes(col = continent)) +
  scale_x_log10() 
```

---

## Data summary w. `geom_smooth()`

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp, size = pop)) +  
  geom_point(aes(col = continent)) + 
  scale_x_log10() +
* geom_smooth()
```

---

## Linear version w. `geom_smooth()`

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp, size = pop)) +  
  geom_point(aes(col = continent)) + 
  scale_x_log10() +
* geom_smooth(method = lm) # lm means "linear model"
```

???

Anything after # (on the same line) is a "comment" and is ignored by R. This is useful for explaining to humans what is going on in the code.

---

## Small multiples: `facet_wrap()`

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp, size = pop)) +  
  geom_point() + 
  scale_x_log10() +
  geom_smooth(method = lm) +
* facet_wrap(vars(continent))
```

---

# Other geoms

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp)) +  
* geom_density2d() +
  scale_x_log10()
```

---

# Other geoms

```r
ggplot(data = gapminder_2007, 
       mapping = aes(x = gdpPercap, y = lifeExp)) +  
* geom_label(aes(label = country)) +
  scale_x_log10()
```

---

# How to learn more

Practice and experiment. (And do problem sets.)

Resources:

- *R For Data Science*
- RStudio primers
- RStudio "Data Visualization" cheat sheet
- Google
- StackOverflow

---

class: bg-full
background-image: url("data:image/png;base64,#assets/data_viz_cheatsheet.png")
background-position: center
background-size: contain

???

Source: [RStudio Cheatsheets](https://www.rstudio.com/resources/cheatsheets/)

---

# Back to the big picture

Components of a `ggplot`:

- data, with observations in rows, attributes in columns
- mapping of attributes to aesthetics (x, y, size, shape, color, transparency, etc)
- geometric objects (`geom`s)

Next: getting data into the right format for plotting (and analysis).

---

# Assignment

Before next lecture, RStudio Cloud primers:

- "Work with data" (https://rstudio.cloud/learn/primers/2): `tibble`, `select()`, `filter()`, `arrange()`, `%>%`, `summarize()`, `group_by()`, `mutate()`
- "Data vizualization" (https://rstudio.cloud/learn/primers/3): more practice with `ggplot`

Alternatively, Chapters 5&6 of R4DS.